You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
However, fixing the discriminator would guarantee degradation, as shown in this ICML paper. https://arxiv.org/pdf/2002.09124.pdf …
Our #icml paper delves deeper into role of dynamic interaction in GAN. Simultaneous update is key to good performance and we further enhance this with use of competitive gradient descent. @Kay12400259 blog: https://f-t-s.github.io/projects/icr/
Very interesting perspective on GAN. Just curious, how expensive CGD would be in GAN training since it seems to require gradient descent on the Hessians? If the purpose is to prevent discriminator from overfitting, wouldn't there be computationally less expensive ways?
Current Pytorch implementation requires 7-9 times more gradient calls, depending on network structures and datasets. It is computationally expensive but in principle, it can further speed up if Pytorch can support both forward and backward mode auto differentiation.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
